Planning to Evaluate a Drug Court? What Are Some Challenges?

Randomized controlled trials are the standard for drug court evaluations.

The most successful drug court evaluations employ experimental designs that randomly assign candidates to drug court programs or to control groups. This allows evaluators to compare outcomes for drug court participants with those of eligible nonparticipants, while controlling for a variety of internal and external factors that might influence any observed differences. When random assignment is not possible, evaluators have employed a variety of methods to establish comparison groups, including using offenders who enrolled for participation in the drug court but did not follow through, those who were refused participation or refused to participate after being invited to do so, those adjudicated before implementation of the drug court being evaluated, or those sentenced to probation prior to implementation of the drug court in question.

The lack of drug court information systems makes demonstrating the equivalence of comparison groups difficult.

When random assignment is not possible, the evaluator must establish the equivalence of the drug court and comparison groups. Inadequate drug court information systems make it difficult to obtain relevant substance abuse history and other demographic data required to assess differences between offenders in the two groups. This lack of data also makes it difficult to assess whether attrition is random in the drug court and control or comparison groups.

Multiyear assessments allow evaluators to determine the impacts of changes in drug courts over time.

Because drug courts are dynamic, many evaluations suffer from the fact that changes are made while the evaluation is ongoing. Ideally the evaluation should be extended to allow enough time for the impacts of the changes to be assessed. Extended evaluations also make it possible to assess the long-term impacts of drug court participation on outcomes such as recidivism.

| | |
| | |